This is a revised application to continue and extend our long-term objective to develop receptor-binding radiotracers for SPECT imaging of human brain. We propose a major new research direction by developing m2 subtype-selective receptor binding radioligands which can be applied to the in vivo study of regional concentrations of the m2 muscarinic receptor (mAChR) subtype in normal controls and in patients with CNS diseases. It is now apparent that quantitative detection of changes of cerebral concentrations of mAChR in Alzheimer's disease (AD) will require selectively imaging the m2 mAChR subtype, currently not possible with available radioligands. Therefore, we propose developing m2-selective radioligands based on the structure of AQ-RA 741, a potent m2-selective antagonist. Several alternative synthetic schemes are proposed to yield smaller and more lipophilic analogues of the parent compound that will be better able to cross the BBB. Each novel compound will be evaluated by apparent binding affinity and subtype-selectivity using a series of cell lines and tissues that each express a single mAChR subtype. Potent and subtype-selective compounds will be further evaluated for their ability to cross the BBB of mice using ex vivo binding studies, and if successful, will be radioiodinated and used further for biodistribution and pharmacokinetic studies in rats. We will perform pharmacokinetic modeling studies to determine whether the regional brain localization of the novel m2-selective radioligands will accurately reflect the regional brain m2 receptor concentration. This determination is required in order to validate that the regional m2 deficits expected in AD can be detectable in terms of a reduced regional radioligand localization. We will apply computer simulation studies of the SPECT, imaging process to a realistic model of the distribution of radioligand bound to m2 receptors in the brain. We will develop -an algorithm which can compensate for the distortions which we demonstrate in the SPECT imaging process, so that we will have the capability of performing quantitatively accurate SPECT neuroreceptor imaging. The proposed studies are designed to validate the quantitative potential of SPECT in studying CNS disorders, and to yield new information about disease states in living humans.